Endotoxin Induced Chorioamnionitis Prevents Intestinal ... · (Invitrogen, San Francisco, CA);...
Transcript of Endotoxin Induced Chorioamnionitis Prevents Intestinal ... · (Invitrogen, San Francisco, CA);...
Endotoxin Induced Chorioamnionitis Prevents IntestinalDevelopment during Gestation in Fetal SheepTim G. A. M. Wolfs1, Wim A. Buurman1, Bea Zoer2, Rob M. J. Moonen2, Joep P. M. Derikx1, Geertje
Thuijls1, Eduardo Villamor2, Markus Gantert3, Yves Garnier3, Luc J. I. Zimmermann2, Boris W. Kramer2*
1 Department of Surgery, Nutrition and Toxicology Research Institute Maastricht (NUTRIM), Maastricht University Medical Centre (MUMC), Maastricht, the Netherlands,
2 Department of Pediatrics, Maastricht University Medical Center (MUMC), School of Oncology and Developmental Biology (GROW), Maastricht, the Netherlands,
3 Department of Obstetrics & Gynecology, University of Cologne, Cologne, Germany
Abstract
Chorioamnionitis is the most significant source of prenatal inflammation and preterm delivery. Prematurity and prenatalinflammation are associated with compromised postnatal developmental outcomes, of the intestinal immune defence, gutbarrier function and the vascular system. We developed a sheep model to study how the antenatal development of the gutwas affected by gestation and/or by endotoxin induced chorioamnionitis. Chorioamnionitis was induced at differentgestational ages (GA). Animals were sacrificed at low GA after 2d or 14d exposure to chorioamnionitis. Long term effects of30d exposure to chorioamnionitis were studied in near term animals after induction of chorioamnionitis. The cellulardistribution of tight junction protein ZO-1 was shown to be underdeveloped at low GA whereas endotoxin inducedchorioamnionitis prevented the maturation of tight junctions during later gestation. Endotoxin induced chorioamnionitisdid not induce an early (2d) inflammatory response in the gut in preterm animals. However, 14d after endotoxinadministration preterm animals had increased numbers of T-lymphocytes, myeloperoxidase-positive cells and gammadeltaT-cells which lasted till 30d after induction of chorioamnionitis in then near term animals. At early GA, low intestinal TLR-4and MD-2 mRNA levels were detected which were further down regulated during endotoxin-induced chorioamnionitis.Predisposition to organ injury by ischemia was assessed by the vascular function of third-generation mesenteric arteries.Endotoxin-exposed animals of low GA had increased contractile response to the thromboxane A2 mimetic U46619 andreduced endothelium-dependent relaxation in responses to acetylcholine. The administration of a nitric oxide (NO) donorcompletely restored endothelial dysfunction suggesting reduced NO bioavailability which was not due to low expression ofendothelial nitric oxide synthase. Our results indicate that the distribution of the tight junctional protein ZO-1, theimmune defence and vascular function are immature at low GA and are further compromised by endotoxin-inducedchorioamnionitis. This study suggests that both prematurity and inflammation in utero disturb fetal gut development,potentially predisposing to postnatal intestinal pathology.
Citation: Wolfs TGAM, Buurman WA, Zoer B, Moonen RMJ, Derikx JPM, et al. (2009) Endotoxin Induced Chorioamnionitis Prevents Intestinal Development duringGestation in Fetal Sheep. PLoS ONE 4(6): e5837. doi:10.1371/journal.pone.0005837
Editor: Nick Gay, University of Cambridge, United Kingdom
Received February 3, 2009; Accepted May 12, 2009; Published June 8, 2009
Copyright: � 2009 Wolfs et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permitsunrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: Grant support: Growth Research Institute, University of Maastricht, Stichting Bevordering Kindergeneeskunde, Maastricht, and grant 016.096.141 of theDutch Research Society to BWK. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: [email protected]
Introduction
Chorioamnionitis is a bacterial infection of the amniotic fluid,
placenta and membranes that is very frequently diagnosed after
preterm birth [1]. The disorder is usually clinically silent which
makes estimates about the onset of infection during gestation
difficult [2]. Chorioamnionitis results from the fetal exposure to
bacteria and bacterial toxins in the contaminated amniotic fluid.
Animal studies showed that fetal aspiration of contaminated
amniotic fluid induces lung inflammation, resulting in profound
changes of the pulmonary homeostasis as indicated by increases in
surfactant lipids and transient injury to the microvasculature and
alveolar septa [3–6]. These alterations in lung development are
associated with increased risk of bronchopulmonary dysplasia [7].
Swallowing of the bacterial contaminated amniotic fluid is likely
to result in antenatal exposure of the premature intestine to
bacteria and their toxins in utero. Whether this exposure results in
inflammatory and or developmentary changes of the gut of
preterm babies is subject of the current study. We hypothesized
that prematurity and inflammation, either alone or combined,
disturb maturation of the gut barrier, the innate immune defense
and vascular function thereby potentially increasing the risk to
postnatal intestinal pathologic conditions. To test our hypothesis,
chorioamnionitis was induced by intraamniotic endotoxin injec-
tion at low gestational age (GA) and animals were delivered either
preterm or near term. In this model, the distribution of the tight
junctional protein zonula occludens protein 1 (ZO-1) was
determined since no data exists on its expression during gestation.
Besides the TLR4 and MD-2 mRNA expression, also the
distribution of either myeloperoxidase (MPO) expressing cells
(polymorphonuclear leukocytes (PMN) as marker for early
inflammation) or T-lymphocytes and gammadelta T-cells (cells
known to be crucial in monitoring and maintaining integrity of
epithelial tissues; late inflammation) were assessed. Finally, the
contraction or relaxation of third-generation mesenteric arteries
was studied. Relaxation was studied in the absence and presence of
PLoS ONE | www.plosone.org 1 June 2009 | Volume 4 | Issue 6 | e5837
a NO donor and concomitantly the expression of endothelial nitric
oxide synthase (eNOS) was assessed.
In this study, we show that a low gestational age and endotoxin
induced chorioamnionitis both prevent maturation of the intestinal
barrier function, the innate immune defense and vascular function.
Materials and Methods
AnimalsThis study was performed according to the guidelines of the
Animal Care Committee of the University of Maastricht, which
approved the protocol. The study was conducted after approval
and in compliance with the guidelines of the ethical committee.
Time-mated Texel ewes with singleton fetuses were randomly
assigned to groups of four or five animals, to receive a single dose
of 10 mg endotoxin (Escherichia coli 055:B5; Sigma Chemical, St.
Louis, MO) resuspended in saline or the equivalent volume of
saline for control animals by ultrasound guided intraamniotic
injections (Figure 1). Chorioamnionitis was induced at 110–111d
GA and at 123d GA. Animals were sacrificed at low GA of 125d
GA after 2d or 14d exposure to chorioamnionitis. The low GA of
125d GA is comparable with a human GA of approximately
27 weeks. Long term effects of 30d exposure to endotoxin were
studied in near term animals at 140d GA (term gestation being
147d GA) after induction of chorioamnionitis at 110–111d
(Figure 1).
All animals were delivered by caesarean section. Lambs
received a lethal i.v. injection of phenobarbiturate. The stomach
content was collected for analysis. Samples of the small intestine
were collected for snap freezing and fixation. Third generation
mesenteric arteries of the small intestine were prepared for analysis
of vascular reactivity.
Antibodies and reagentsIntestinal tissue was analysed with the following antibodies:
monoclonal antibody against endothelial nitric oxide synthase
(eNOS) (BD Transduction laboratories, San Jose, CA); rabbit
antibodies against human MPO and CD3 Dakocytomation
(Glostrup, Denmark) or Zonula Occludens protein 1 (ZO-1)
(Invitrogen, San Francisco, CA); monoclonal antibody IL-A29 for
the detection of gammadelta T-cells (VMRD, Pullman, WA).
Secondary antibodies, biotin conjugated rabbit anti-mouse and
Texas red conjugated goat anti-rabbit were purchased from
Dakocytomation and peroxidase conjugated goat anti-rabbit from
Jackson (West Grove, PA).
ImmunohistochemistrySmall intestinal biopsies, fixed overnight in 4% formaldehyde,
were embedded in paraffin and sectioned at 3 mm. For staining of
CD-3 and eNOS expressing cells, slides were heated in 10 mM
Na-citrate (pH 6.0) for 20 min. Endogenous peroxidase activity
was blocked with 0.3% H2O2 in methanol and subsequently, slides
were blocked with either normal goat serum (eNOS and MPO) or
bovine serum albumin (CD-3) for 30 min at room temperature.
Slides were incubated with the primary antibody of interest for 1 h
at RT (CD-3 and MPO) or overnight at 4uC (eNOS). After
washing, sections were incubated with the appropriate secondary
conjugated antibody. Binding of the eNOS antibody was
demonstrated by the streptavidin-biotin system (Dakocytomation)
and binding of antibodies against CD-3 and MPO was detected
using a peroxidase conjugated secondary antibody. Positive
staining was visualized by applying 3-amino-9-ethylcarbazole
(AEC, Sigma); nuclei were counterstained with haematoxylin.
The number of cells exhibiting immunostaining for MPO and
CD-3 were counted per high power field (200x). Figures were
Figure 1. Experimental design. Gestational development and effects of chorioamnionitis were studied at two different gestational ages. At 125dGA, fetuses were comparable to 27 weeks of human gestation. Lambs were almost near term at 140d GA since term gestation is 147d.Chorioamnionitis was induced by a single injection of endotoxin under ultrasound guidance at 111d GA and at 123d GA. Animals were delivered at125d GA and animals of the control group underwent the same procedure with an injection of saline. Animals of 140d GA had a separate controlgroup to assess gestational changes.doi:10.1371/journal.pone.0005837.g001
Gestation and Chorioamnionitis
PLoS ONE | www.plosone.org 2 June 2009 | Volume 4 | Issue 6 | e5837
expressed as number of MPO or CD-3 expressing cells per high
power field.
Frozen tissue sections were first fixed in filtered, ice-cold acetone
and next in 4% paraformaldehyde. Endogenous peroxidase was
blocked with 0.3% H2O2 in methanol. After blocking with 10%
normal goat serum, sections were incubated for 2 h with a mab
against gamma delta T-cells. After washing, an appropriate biotin
conjugated secondary antibody was used. Binding of the primary
antibody was demonstrated by the streptavidin-biotin system
(Dakocytomation) and visualized by AEC. Nuclei were counter-
stained with haematoxylin. All slides were photographed by a
Nikon eclipse E800 microscope with a Nikon digital camera
DXM1200F and were analysed by the computer program, Lucia
G color image analysis.
ImmunofluoresenceTight junction distribution was examined by immunofluores-
cent staining of frozen sections (3 mm) with an antibody to Zonula
Occludens protein 1 (ZO-1). Ileum sections were fixed with 4%
paraformaldehyde. After blockage of non-specific binding sites
with 10% goat serum, slides were incubated with anti-ZO-1 for
1 h. Thereafter, the sections were incubated 45 min with Texas
red conjugated goat anti-rabbit antibody, followed by 2 min
incubation with 49,6-diamino-2-phenyl indole (DAPI), dehydrated
in ascending ethanol series and mounted in Fluorescence
Mounting Solution (Dakocytomation). The distribution of ZO-1
was recorded at a magnification of 200x using the Metasystems
Image Pro System (black and white charge-couple device camera;
Metasystems, Sandhausen, Germany) mounted on a Leica DM-
RE fluorescence microscope (Leica, Wetzler, Germany). All
images were taken at equal time-exposures after being normalized
to negative control sections without primary antibody, to exclude
non-specific binding of the secondary antibody or autofluores-
cence. At least 25 microscopic fields for each tissue section were
studied.
Evaluation of mRNA levels by RT-PCRFor RT-PCR total RNA was extracted from ileal tissues using
the SV Total RNA isolation system (Promega, Madison, WI).
Next, isolated total RNA was treated with RQ1 RNase-Free
DNase (Promega) and reverse transcribed using oligo (dT) primer
and Moloney murine leukemia virus reverse transcriptase (Life
technologies, Paisley, United Kingdom) according to the supplier’s
recommendations. cDNA samples were standardized based on the
content of GAPDH cDNA as housekeeping gene. GAPDH cDNA
was evaluated by performance of a GAPDH PCR on multiple
dilutions of each cDNA sample. The amount of amplified product
was estimated by densitometry of ethidiumbromide stained 1.2%
agarose gels using a CCD camera and Imagemaster VDS software
(Amersham, Uppsala, Sweden). Primer sequences for GAPDH
were:59-CGT CTT CAC CAC CAT GGA GA-39 (sense primer)
and 59-CGG CCA TCA CGC CAC AGT TT–39 (antisense
primer); primers used for the amplification of TLR4 mRNA: 59-
CCT CTC CAC CTT GAT ACT GAC G -39 (sense primer) and
59- GGT ACC TGG TTC AAT AAA GCC T-39 (antisense
primer) and primers designed for amplification of MD-2 mRNA:
59- CCT GTT TTC TTC CAT ATT TAC TG (sense primer)
and 59- AAT AAC TTC TTT GCG CTT TGG–39. PCR
reactions with GAPDH, TLR4 or MD-2 specific primers were
performed using appropriate dilutions of the cDNA. PCR
reactions were performed in a total volume of 25 ml in PCR
buffer (Perkin Elmer/Cetus, Emeryville, CA) in the presence of
0.2 mM dNTP (Amersham), 1.0 mM of each primer, 0.3 mM
MgCl2 and 0.5 U Taq polymerase (Perkin Elmer). PCR conditions
for each primer couple were as follows: GAPDH: 95uC for 30 sec,
55uC for 30 sec and 72uC for 30 sec during 21 cycles; TLR4:
95uC for 30 sec, 54uC for 30 sec and 72uC for 30 sec during 40
cycles; MD-2: 95uC for 30 sec, 52uC for 30 sec and 72uC for
30 sec during 33 cycles. Levels of TLR4 and MD-2 RNA
expression were evaluated by densitometric image analysis as
described above. Relative TLR4 and MD-2 mRNA levels were
calculated by comparison of band intensities of the RT-PCR
products to standard curves prepared by PCR amplifications on
dilution series of a highly concentrated ileal ovine cDNA.
Arterial reactivitySince ischemia as a consequence of vascular compromise is one
of the frequently proposed risk factors for impaired intestinal
homeostasis, responsiveness to contractile and endothelium-
dependent and -independent relaxant agonists was assessed in
isolated third-generation mesenteric arteries. Artery rings were
mounted in a wire myograph, bathed in Krebs solution (gassed
with 95% air/5% CO2, pH 7.4), and normalized to a resting
pretension corresponding to an intraluminal pressure of 40 mm
Hg (125d GA) or 50 mm Hg (140d GA). The contractions induced
by KCl (62.5 mM), the thromboxane A2 mimetic U46619 (10210–
1027M) and the adrenergic receptor agonist norepinephrine (NE;
10210–1024M) was assessed. In addition, the endothelium-
dependent relaxation (evoked by acetylcholine, Ach;1029–
1024M) [8] and endothelium–independent relaxation evoked by
the NO donor sodium nitroprusside (SNP; 1029–1024M) were
analyzed. For the study of the relaxant responses, vessels were pre-
contracted with U46619 (1027M) as previously described [9].
Statistical analysisThe number of cells exhibiting immunostaining for MPO and
CD-3 were counted per high power field. A Mann-Whitney U-test
was used for between-group comparisons. Vascular contractile and
relaxation responses are represented as mean6SEM and the 2-
way unpaired t-test was used to test for significant differences
between saline and endotoxin treated groups. TLR4 and MD-2
mRNA levels were represented as mean6SEM and differences
between preterm and term lambs or saline and endotoxin injected
animals were compared using an unpaired t-test. Statistical
calculations were made using SPSS 15.0 for Windows (SPSS,
Chicago, IL) and differences were considered statistically signifi-
cant at p,0.05.
Results
Tight junctionsTight junctions have an important role in maintaining the
barrier integrity by connecting enterocytes [10,11]. In order to
analyse the maturation and the influence of endotoxin in the fetal
gut on this essential component of the gut barrier, the distribution
of the zonula occludens protein 1 (ZO-1) was determined. ZO-1
staining was fragmented in premature saline treated control
animals of 125d GA (Figure 2A). Interestingly, endotoxin exposure
for 2d did not change the pattern (Figure 2B). Fetuses exposed to
endotoxin for 14d showed further fragmented tight junctional
distribution at 125d GA (Figure 2C). In contrast, in near term
saline injected animals of 140d GA, ZO-1 proteins formed a
continuous ring, which is indicative for a normal tight junctional
distribution (Figure 2D). Interestingly, in these near term lambs of
140d GA, exposure to endotoxin for 30d resulted in partial loss of
the ZO-1 protein in comparison to the age-matched control
animals with saline injection (Figure 2E).
Gestation and Chorioamnionitis
PLoS ONE | www.plosone.org 3 June 2009 | Volume 4 | Issue 6 | e5837
Gestation and Chorioamnionitis
PLoS ONE | www.plosone.org 4 June 2009 | Volume 4 | Issue 6 | e5837
Inflammatory parametersIn preterm control animals of 125d GA, hardly any MPO
expressing cells were detected (data not shown). At this GA,
endotoxin exposure for 2d did not result in a significant increase of
infiltrating intestinal MPO positive cells (data not shown).
Interestingly, at 14 days post endotoxin treatment, large numbers
of MPO positive cells were present in the lamina propria of the
immature intestine (Figure 3A). Remarkably, 30d post endotoxin
treatment, the number of MPO expressing cells reactivity was still
significantly enhanced in the near term fetus (Figure 3B) when
compared to saline treated control animals of this GA (Figure 3C).
For each individual animal, the number of infiltrating MPO
expressing cells was counted per high-powerfield and displayed in
a scatter plot (Figure 3D).
The recruitment of CD-3 positive T-cells into the fetal gut
paralleled that of PMN: in preterm lambs exposed to endotoxin
Figure 3. At 14 and 30d after intraamniotic endotoxin injection, massive influx of PMNs was detected (A+B). At 140d GA, some MPOpositive cells were detected in control sections (C). For each experimental group mean cell counts of MPO positive cells are depicted per high-powerfield (D).doi:10.1371/journal.pone.0005837.g003
Figure 2. Immunolocalisation of ZO-1 (red) in the fetal intestine. At 125d GA, a fragmented staining pattern for ZO-1 was seen in controlanimals (A) or lambs exposed to endotoxin for 2 (B) or 14d (C). At 140d GA, a normal ZO-1 distribution was detected in control animals (D) thatbecame disrupted upon endotoxin exposure for 30d (E). Magnification 200x. For inset, 1000x magnification was used.doi:10.1371/journal.pone.0005837.g002
Gestation and Chorioamnionitis
PLoS ONE | www.plosone.org 5 June 2009 | Volume 4 | Issue 6 | e5837
for 2d, no significant increase of CD-3 expressing T-cells in the
lamina propria was observed when compared with time matched
control tissues of 125d GA (data not shown). In contrast, increased
numbers of CD-3 expressing cells were seen at 14d and 30d post
endotoxin exposure in both preterm and near term animals of 125
and 140d GA respectively, with the highest density in the 14d
group (Figure 4A+B). For each individual section, the number of
CD-3 positive cells was quantified per high-powerfield and data
are depicted in a scatter plot (Figure 4C).
Next, the distribution of gammadelta T-cells was evaluated, cells
known to be crucial in monitoring and maintaining integrity of
epithelial tissues [12]. In saline treated preterm animals of 125d
GA, hardly any gammadelta T-cells were present in the lamina
propria or lower mucosa (data not shown). Accordingly, in these
preterm lambs, migration of gammadelta T-cells from the lower to
the upper mucosa occurred sporadically after 2 or 14d post
endotoxin treatment (Figure 5A+B). In saline injected animals of
140d GA, gammadelta T-cells were also undetectable in the
lamina propria (Figure 5C). However, in these near term lambs,
the lymphoid follicles in the lower mucosa became populated with
gammadelta T-cells (Figure 5D). Interestingly, 30 d of intraam-
niotic endotoxin exposure in near term animals resulted in
Figure 4. Intraamniotic endotoxin results in increased ileal T-cell density. Significant increase of CD-3 immunoreactivity was seen at 14 and30d after endotoxin exposure (A+B). For each experimental group mean cell counts of CD-3 positive cells are depicted per high-power field (C).doi:10.1371/journal.pone.0005837.g004
Gestation and Chorioamnionitis
PLoS ONE | www.plosone.org 6 June 2009 | Volume 4 | Issue 6 | e5837
Figure 5. Endotoxin induced chorioamnionitis results in migration of gammadelta T-cells to sites of mucosal damage at late GA.After 2d or 14d of endotoxin exposure, hardly any gammadelta T-cells were detected in the upper mucosa of the preterm intestine (A+B). At 140d GA,gammadelta T-cells were not detected in the lamina propria whereas lymphoid follicles in the basal mucosa became populated with gamma delta T-cells (C+D). 30d after endotoxin injection, increased numbers of gammadelta T-cells were found within areas of mucosal damage (E).doi:10.1371/journal.pone.0005837.g005
Gestation and Chorioamnionitis
PLoS ONE | www.plosone.org 7 June 2009 | Volume 4 | Issue 6 | e5837
migration of gammadelta T-cells from the lower to the upper
mucosa in close proximity to areas of damaged intestinal epithelial
cells (Figure 5E).
TLR4 and MD-2 expressionThe molecular complex responsible for the recognition and
signalling of endotoxin is formed by TLR4 and its indispensable
cofactor MD-2 [13,14]. In preterm lambs of 125d, weak
constitutive TLR4 and MD-2 mRNA expression was detected
and this expression even decreased within 2d following intraam-
niotic endotoxin treatment (Figure 6). Also intraamniotic exposure
to endotoxin for 14d in preterms resulted in a downregulation of
the TLR4 and MD-2 mRNA levels (Figure 6). In saline treated
near term sheep of 140d GA, TLR4 and MD-2 expression was
significantly increased when compared with preterm controls of
125d GA. However, intraamniotic endotoxin exposure for 30d
also resulted in a strong reduction of the TLR4 and MD-2
expression in near term lambs (Figure 6).
Vascular contraction and relaxation responsesContractions of third-generation mesenteric arteries, evoked by
the thromboxane receptor agonist U46619 increased with
gestational age (Figure 7A). After endotoxin-induced chorioamni-
onitis, the potency and the contractile efficacy of U46619 was
significantly increased at early (125d) but not at late GA (140d)
(Figure 7A). In contrast, the contractions induced by KCl (data not
shown) and NE (Figure 7B) were not statistically different, both in
control animals at different gestational ages and in lambs exposed
to chorioamnionitis.
Either ACh (Figure 7C) or SNP (Figure 7D) relaxed U46619-
contracted fetal mesenteric arteries in a concentration-dependent
manner and these relaxations did not change with gestational age.
Interestingly, ACh-evoked relaxation was almost abolished in the
125d fetuses exposed to endotoxin (Figure 7C). However this effect
of endotoxin on ACh-induced relaxation was not further observed
at 140d GA (Figure 7C).
Interestingly, the presence of SNP completely reversed the
impaired relaxation of premature animals exposed to endotoxin
(Figure 7D). These findings suggest a vascular impairment during
chorioamnionitis at low GA on account of reduced release of NO
by endothelial cells.
eNOS expressionSince the findings described above point at endothelial
dysfunction due to impaired synthesis of NO, the distribution of
endothelial nitric oxide synthase (eNOS) was studied in mesenteric
arteries. Constitutive eNOS expression was absent in preterm and
near term control tissues of 125 and 140d GA respectively (data
not shown). However, 2 days after intraamniotic endotoxin
injection in preterm animals, eNOS was weakly expressed by
submucosal vascular endothelia (Figure 8A). At this GA,
endothelial staining further enhanced at 14 days after endotoxin
exposure (Figure 8B). eNOS immunoreactivity was also present in
the vascular endothelium of the submucosa in near term animals
that were exposed to endotoxin for 30d (Figure 8C).
Discussion
This study shows for the first time the effects of gestation either
alone or combined with antenatal exposure to bacterial toxins on
the development of the fetal gut. A sheep model was used that
closely resembles the human situation of preterm babies that are in
utero exposed to chorioamnionitis by bacterial toxins resulting in
adverse outcomes [15]. Administration of endotoxin to the
amniotic fluid resulted in increased concentrations of endotoxin
activity in the stomach fluid of animals exposed for 2d to
endotoxin (data not shown), indicating passage of endotoxin to the
fetal gut. This experimental set up allowed us to investigate
whether prematurity and inflammation, either alone or combined,
influence the development of the intestine in utero.
To our knowledge, this is the first report showing that the
cellular distribution of tight junction protein ZO-1 was underde-
veloped at 125d GA in sheep. Although ZO-1 is not the only tight
junction protein responsible for paracellular barrier sealing, it
plays a central role in maintenance of paracellular intestinal
barrier integrity by connection of the intercellular tight junction
proteins (claudins and occludin) with the cell cytoskeleton.
The presence of fragmented tight junctions in the gut of preterms
(125d GA) is inherent to immaturity and correlates with diminished
gut barrier function in preterm infants, as assessed by sugar
absorption tests [16]. In near term control tissue of 140d GA,
staining of ZO-1 formed a continuous band lining the mucosal
epithelium, indicating maturation of the gut barrier during gestation.
This study shows that bacterial toxins impair maturation of tight
junctions during gestation, an effect which lasted as long as 30d post
endotoxin administration. These data are in line with earlier in vitro
and in vivo findings showing that the permeability of the intestinal
epithelium is impaired after exposure to LPS [17–19]. An imperfect
tight junctional distribution, either due to immaturity or inflam-
mation suggests an easy access of microbial toxins to the mucosa
and the inner layers of the gut, both in utero and after birth. This led
us to investigate the mRNA levels of the molecules representing the
endotoxin recognition complex during gestation and after intraam-
niotic endotoxin injection. A gestational dependent increase of
intestinal TLR4 and MD-2 mRNA was detected. In line, recent
findings from our group showed that the human MD-2 protein is
expressed in term but absent in preterm infants [20]. Combined,
these findings suggest that the capacity to recognize and sense
Gram-negative derived endotoxin by the fetal intestine is immature
at low GA. Interestingly, within 48 h after intraamniotic endotoxin,
TLR4 mRNA became undetectable and concomitantly, also MD-2
expression levels were reduced. The nature of this rapid
downregulation needs further investigation [21,22], but it might
explain the here observed hyporesponsiveness of the fetal gut to
endotoxin at low GA. In particular, an impaired recruitment of
MPO-positive cells or T lymphocytes was detected in the intestine at
2d after intraamniotic injection of endotoxin at 125d GA, likely due
to absence of endotoxin recognition. This is consistent with earlier
studies in this chorioamnionitis model showing that proinflamma-
tory cytokine mRNAs were unaltered in the fetal intestine, from five
hours to 25d after intraamniotic injection of endotoxin [23]. In
contrast to the intestinal TLR4 and MD-2 mRNA expression data
presented here, previous studies with this animal model showed that
intraamniotic endotoxin induced pulmonary TLR4 mRNA expres-
sion within 2d [24]. In line, endotoxin induced chorioamnionitis
provoked a rapid inflammatory reaction in the respiratory system
within 24 hours [5,6,25,26].
Interestingly, at 14 and even 30d after intraamniotic endotoxin
injection, a strong intestinal inflammatory response was present
with a massive influx of MPO positive cells and T-lymphocytes.
Paradoxically, at 14 and 30d post endotoxin exposure, TLR4
mRNA was still undetectable while MD-2 expression levels
remained reduced. Whether this sustained inflammatory process
is the result of a broken or absent tolerance to endotoxin remains
to be elucidated [22,27]. In addition, activation via TLR4-MD-2
independent intestinal innate immune receptors cannot be
excluded with these experiments. Alternatively, the delayed
inflammatory response in the fetal gut might be the consequence
Gestation and Chorioamnionitis
PLoS ONE | www.plosone.org 8 June 2009 | Volume 4 | Issue 6 | e5837
of a systemic rather than a local response of the fetal intestine.
Esophageal occlusion experiments need to be performed to define
the site of stimulation of the massive influx of PMNs and T-cells in
the fetal gut.
At 30d post endotoxin exposure, inflammatory parameters were
decreased compared to the 14d group, and this was paralleled by
migration of gammadelta T-cells from the lower mucosa towards
compromised epithelial surfaces. These findings are supported by
Figure 6. Ileal TLR4 and MD-2 mRNA was evaluated by ovine specific RT-PCR amplification of cDNA samples from healthy controllambs and lambs subjected to endotoxin induced chorioamnionitis. Representative DNA fragments are shown for each group (n = 4). cDNAswere standardized for GAPDH content. Quantitative data were obtained by densitometric evaluation of RT-PCR products which were compared to astandard curve obtained by amplification of a serial dilution of highly concentrated cDNA. At 125d GA, TLR4 and MD-2 mRNA levels were significantly(*) lower compared to animals at 140d GA. TLR4 and MD-2 mRNA expression was significantly reduced following intraamniotic endotoxin injection for2, 14 and 30d.doi:10.1371/journal.pone.0005837.g006
Gestation and Chorioamnionitis
PLoS ONE | www.plosone.org 9 June 2009 | Volume 4 | Issue 6 | e5837
Gestation and Chorioamnionitis
PLoS ONE | www.plosone.org 10 June 2009 | Volume 4 | Issue 6 | e5837
a report from Chen et al. who showed that gammadelta T-cells
preserve the integrity of stressed or injured intestinal mucosa [12].
Moreover, our study showed that the lymphoid follicles of the
healthy fetal intestine became populated with gammadelta T-
lymphocytes between 125 days and 140 days of gestational age. In
addition to the fragmented tight junctional distribution as seen in
preterms or following endotoxin treatment, the absence of gamma
delta T-cells in the mucosa at 125d GA suggests a reduced
capacity to preserve the already compromised epithelium.
This study shows that intrauterine exposure to endotoxin induced
a transient increase in the mesenteric vascular responsiveness to the
thromboxane receptor agonist U46619. In addition a profound
suppression of ACh-induced endothelium-dependent relaxation in
mesenterial arteries was found during chorioamnionitis at low GA a
process that was completely reversed by administration of the NO
donor SNP. Our results strongly suggest the presence of an
endotoxin-evoked transient endothelial damage and it could be
speculated that a similar mechanism of endothelial dysfunction and
diminished NO production might be partially responsible for the
vascular vulnerability of the intestine in preterm infants [28].
The decreased NO availability could have several explanations.
First, clinical reports showed that the intestinal synthesis of the
Figure 8. Intraamniotic endotoxin induces eNOS expression. Weak eNOS staining was seen at 2d post endotoxin treatment (A) whileexpression increased in animals exposed to endotoxin for 14 or days 30d (B-C).doi:10.1371/journal.pone.0005837.g008
Figure 7. Thromboxane A2 mimic U46619 induced contractions of jejunal arteries increased with gestational age and weresignificantly (*) (P,0.05) augmented after intra-amniotic endotoxin injection in the 125d (P,0.05) but not in the 140d GA animals(A). Concentration-dependent contractile effects of norepinephrine tend to increase (p.0.05) during gestation but remained unaltered duringchorioamnionitis (B). Concentration-dependent relaxation of mesenteric arteries induced by acetylcholine was unaltered during gestation and wassignificantly (*) inhibited at 14d but not after 30d of endotoxin exposure (C). The relaxation response curve for sodium nitroprusside remainedidentical during gestation or chorioamnionitis (D).doi:10.1371/journal.pone.0005837.g007
Gestation and Chorioamnionitis
PLoS ONE | www.plosone.org 11 June 2009 | Volume 4 | Issue 6 | e5837
important NO precursor arginine is strongly reduced in premature
infants [29–32]. Second, a mechanism for the reduced NO
biosynthesis would be a decreased expression of eNOS. Paradox-
ically, our data showed an enhanced expression of eNOS in the
chorioamnionitis exposed animals. Such an increase in eNOS
expression indicates the presence of a compensatory mechanism,
which would attempt to maintain normal NO level by upregulat-
ing eNOS [33,34].
In summary, in this study evidence is provided that both
prematurity and chorioamnionitis are associated with impaired
development of the intestinal innate immune defence, the tight
junctional distribution and vascular function in utero. Although at
present, we can only speculate about the postnatal consequences of
these developmental disorders, immaturity of the parameters
investigated in this study are all associated with neonatal intestinal
pathology and NEC in particular [35–38]. The immature gut of
preterm babies is a risk factor on its own which is further
aggravated by the presence of antenatal inflammation. In addition,
reduced expression of innate immune receptors as seen in the
intestine of preterm lambs suggests that the capacity to recognize
and sense Gram-negative derived endotoxin by the fetal intestine
is immature at low GA. Such a lack of endotoxin sensing in the
preterm intestine, against the background of a disrupted tight
junctional distribution, supports the theory that an inadequate
immune response contributes to the pathogenesis of NEC,
potentially by allowing bacterial invasion and subsequent
overgrowth in premature infants. Further research is needed to
assess the importance of preterm delivery and the consequences of
inflammation in utero on the development of neonatal intestinal
pathologies such as NEC.
Acknowledgments
We thank Leon Janssen, Isabelle Meijssen, Nona Jongmans and Verena
Lambermont for the excellent technical help.
Author Contributions
Conceived and designed the experiments: TGAMW WB LJIZ BWK.
Performed the experiments: TGAMW BZ RMJM JD GT BWK. Analyzed
the data: TGAMW WB BZ RMJM JD GT EV MG YG LJIZ BWK.
Contributed reagents/materials/analysis tools: EV MG YG. Wrote the
paper: TGAMW WB EV LJIZ BWK.
References
1. Goldenberg RL, Hauth JC, Andrews WW (2000) Intrauterine infection andpretermdelivery. N Engl J Med 342: 1500–1507.
2. Saji F, Samejima Y, Kamiura S, Sawai K, Shimoya K, et al. (2000) Cytokine
production in chorioamnionitis. J Reprod Immunol 47: 185–196.3. Kramer BW, Kramer S, Ikegami M, Jobe AH (2002) Injury, inflammation, and
remodeling in fetal sheep lung after intra-amniotic endotoxin. Am J PhysiolLung Cell Mol Physiol 283: L452–459.
4. Kallapur SG, Jobe AH (2006) Contribution of inflammation to lung injury and
development. Arch Dis Child Fetal Neonatal Ed 91: F132–135.5. Kramer BW, Moss TJ, Willet KE, Newnham JP, Sly PD, et al. (2001) Dose and
time response after intraamniotic endotoxin in preterm lambs. Am J Respir CritCare Med 164: 982–988.
6. Kramer BW (2008) Antenatal inflammation and lung injury: prenatal origin ofneonatal disease. J Perinatol 28 Suppl 1: S21–27.
7. Jobe AH (2003) Antenatal factors and the development of bronchopulmonary
dysplasia. Semin Neonatol 8: 9–17.8. Shaul PW, Farrar MA, Zellers TM (1992) Oxygen modulates endothelium-
derived relaxing factor production in fetal pulmonary arteries. Am J Physiol 262:H355–364.
9. Villamor E, Kessels CG, Fischer MA, Bast A, de Mey JG, et al. (2003) Role of
superoxide anion on basal and stimulated nitric oxide activity in neonatal pigletpulmonary vessels. Pediatr Res 54: 372–381.
10. Nusrat A, Parkos CA, Verkade P, Foley CS, Liang TW, et al. (2000) Tightjunctions are membrane microdomains. J Cell Sci 113 (Pt 10): 1771–1781.
11. Viswanathan VK, Hecht G (2000) Innate immunity and the gut. Curr Opin
Gastroenterol 16: 546–551.12. Chen Y, Chou K, Fuchs E, Havran WL, Boismenu R (2002) Protection of the
intestinal mucosa by intraepithelial gamma delta T cells. Proc Natl AcadSci U S A 99: 14338–14343.
13. Beutler B (2004) Inferences, questions and possibilities in Toll-like receptorsignalling. Nature 430: 257–263.
14. Nagai Y, Akashi S, Nagafuku M, Ogata M, Iwakura Y, et al. (2002) Essential
role of MD-2 in LPS responsiveness and TLR4 distribution. Nat Immunol 3:667–672.
15. Been JV, Kramer BW, Zimmermann LJ (2008) In utero and early-life conditionsand adult health and disease. N Engl J Med 359: 1523–1524; author reply 1524.
16. Rouwet EV, Heineman E, Buurman WA, ter Riet G, Ramsay G, et al. (2002)
Intestinal permeability and carrier-mediated monosaccharide absorption inpreterm neonates during the early postnatal period. Pediatr Res 51: 64–70.
17. Spitz J, Yuhan R, Koutsouris A, Blatt C, Alverdy J, et al. (1995)Enteropathogenic Escherichia coli adherence to intestinal epithelial monolayers
diminishes barrier function. Am J Physiol 268: G374–379.18. Philpott DJ, McKay DM, Sherman PM, Perdue MH (1996) Infection of T84
cells with enteropathogenic Escherichia coli alters barrier and transport
functions. Am J Physiol 270: G634–645.19. Hackam DJ, Upperman JS, Grishin A, Ford HR (2005) Disordered enterocyte
signaling and intestinal barrier dysfunction in the pathogenesis of necrotizingenterocolitis. Semin Pediatr Surg 14: 49–57.
20. Wolfs TG, Derikx JP, Hodin CM, Vanderlocht J, Driessen A, et al. (2009) The
localization of the LPS-recognition complex in the human healthy and inflamedpremature and adult gut. Inflamm Bowel Dis, In press.
21. Kramer BW, Ikegami M, Moss TJ, Nitsos I, Newnham JP, et al. (2005)
Endotoxin-induced chorioamnionitis modulates innate immunity of monocytes
in preterm sheep. Am J Respir Crit Care Med 171: 73–77.
22. Otte JM, Cario E, Podolsky DK (2004) Mechanisms of cross hyporesponsiveness
to Toll-like receptor bacterial ligands in intestinal epithelial cells. Gastroenter-
ology 126: 1054–1070.
23. Kallapur SG, Willet KE, Jobe AH, Ikegami M, Bachurski CJ (2001) Intra-
amniotic endotoxin: chorioamnionitis precedes lung maturation in preterm
lambs. Am J Physiol Lung Cell Mol Physiol 280: L527–536.
24. Hillman NH, Moss TJ, Nitsos I, Kramer BW, Bachurski CJ, et al. (2008) Toll-
like receptors and agonist responses in the developing fetal sheep lung. Pediatr
Res 63: 388–393.
25. Moss TJ, Nitsos I, Kramer BW, Ikegami M, Newnham JP, et al. (2002) Intra-
amniotic endotoxin induces lung maturation by direct effects on the developing
respiratory tract in preterm sheep. Am J Obstet Gynecol 187: 1059–1065.
26. Kallapur SG, Jobe AH, Ball MK, Nitsos I, Moss TJ, et al. (2007) Pulmonary and
systemic endotoxin tolerance in preterm fetal sheep exposed to chorioamnionitis.
J Immunol 179: 8491–8499.
27. Kramer BW, Kallapur SG, Moss TJ, Nitsos I, Newnham JP, et al. (2009) Intra-
amniotic LPS modulation of TLR signaling in lung and blood monocytes of fetal
sheep. Innate Immun 15: 101–107.
28. Moonen RM, Paulussen AD, Souren NY, Kessels AG, Rubio-Gozalbo ME, et
al. (2007) Carbamoyl phosphate synthetase polymorphisms as a risk factor for
necrotizing enterocolitis. Pediatr Res 62: 188–190.
29. Wu G, Jaeger LA, Bazer FW, Rhoads JM (2004) Arginine deficiency in preterm
infants: biochemical mechanisms and nutritional implications. J Nutr Biochem
15: 442–451.
30. Zamora SA, Amin HJ, McMillan DD, Kubes P, Fick GH, et al. (1997) Plasma
L-arginine concentrations in premature infants with necrotizing enterocolitis.
J Pediatr 131: 226–232.
31. Becker RM, Wu G, Galanko JA, Chen W, Maynor AR, et al. (2000) Reduced
serum amino acid concentrations in infants with necrotizing enterocolitis.
J Pediatr 137: 785–793.
32. Richir MC, Siroen MP, van Elburg RM, Fetter WP, Quik F, et al. (2007) Low
plasma concentrations of arginine and asymmetric dimethylarginine in
premature infants with necrotizing enterocolitis. Br J Nutr 97: 906–911.
33. Cai H, Harrison DG (2000) Endothelial dysfunction in cardiovascular diseases:
the role of oxidant stress. Circ Res 87: 840–844.
34. Lopez-Ongil S, Hernandez-Perera O, Navarro-Antolin J, Perez de Lema G,
Rodriguez-Puyol M, et al. (1998) Role of reactive oxygen species in the signalling
cascade of cyclosporine A-mediated up-regulation of eNOS in vascular
endothelial cells. Br J Pharmacol 124: 447–454.
35. Lin PW, Stoll BJ (2006) Necrotising enterocolitis. Lancet 368: 1271–1283.
36. Srinivasan PS, Brandler MD, D’Souza A (2008) Necrotizing enterocolitis. Clin
Perinatol 35: 251–272, x.
37. Kawase Y, Ishii T, Arai H, Uga N (2006) Gastrointestinal perforation in very
low-birthweight infants. Pediatr Int 48: 599–603.
38. Boston VE (2006) Necrotising enterocolitis and localised intestinal perforation:
different diseases or ends of a spectrum of pathology. Pediatr Surg Int 22:
477–484.
Gestation and Chorioamnionitis
PLoS ONE | www.plosone.org 12 June 2009 | Volume 4 | Issue 6 | e5837